Association between statin use and ischemic stroke or major hemorrhage in patients taking dabigatran for atrial fibrillation

Tony Antoniou; Erin M. Macdonald; Zhan Yao; Simon Hollands; Tara Gomes; Mina Tadrous; Muhammad M. Mamdani; David N. Juurlink

doi:10.1503/cmaj.160303

Article Figures & Tables

Figures

Tables

Download figure
Open in new tab
Download powerpoint
Figure 1:
Study design. TIA = transient ischemic attack.

Tables

Figures

View popup

Table 1:

Numbers of patients taking each statin, by diagnosis of ischemic stroke or transient ischemic attack, or major hemorrhage

Statin	No. (%)
Statin	Stroke or TIA n = 1985^*	Major hemorrhage n = 5582^*
Atorvastatin	986 (49.7)	2778 (49.8)
Fluvastatin	6 (0.3)	26 (0.5)
Lovastatin	12 (0.6)	27 (0.5)
Pravastatin	61 (3.1)	190 (3.4)
Rosuvastatin	743 (37.4)	2059 (36.9)
Simvastatin	177 (8.9)	502 (9.0)

Note: TIA = transient ischemic attack.
↵* Stroke or TIA: 397 cases + 1588 controls; major hemorrhage: 1117 cases + 4465 controls.

View popup

Table 2:

Characteristics of study patients

Characteristic	Study 1 (stroke or TIA), no. (%)^*			Study 2 (major hemorrhage), no. (%)^*

	Cases n = 397	Controls n = 1588	Standardized difference^†	Cases n = 1117	Controls n = 4465	Standardized difference^†
Age, yr

Median (IQR)	82 (77–86)	81 (77–86)	0.02	82 (77–86)	82 (77–86)	0.03

66–74	74 (18.6)	301 (19.0)	0.01	214 (19.2)	863 (19.3)	0.00

75–84	203 (51.1)	833 (52.5)	0.03	561 (50.2)	2229 (49.9)	0.01

≥ 85	120 (30.2)	454 (28.6)	0.04	342 (30.6)	1373 (30.8)	0.00

Sex, female	209 (52.6)	836 (52.6)	0.00	468 (41.9)	1872 (41.9)	0.00

Stroke or transient ischemic attack in previous 5 years	144 (36.3)	197 (12.4)	0.65	170 (15.2)	598 (13.4)	0.05

Charlson Comorbidity Index score

No hospital admission	109 (27.5)	811 (51.1)	0.48	337 (30.2)	2230 (49.9)	0.40

0	44 (11.1)	219 (13.8)	0.08	159 (14.2)	668 (15.0)	0.02

1	70 (17.6)	223 (14.0)	0.10	188 (16.8)	638 (14.3)	0.07

≥ 2	174 (43.8)	335 (21.1)	0.53	433 (38.8)	929 (20.8)	0.42

History of congestive heart failure	172 (43.3)	586 (36.9)	0.13	579 (51.8)	1664 (37.3)	0.30

History of angina	67 (16.9)	241 (15.2)	0.05	257 (23.0)	704 (15.8)	0.19

History of diabetes	198 (49.9)	693 (43.6)	0.13	528 (47.3)	1906 (42.7)	0.09

History of hypertension	379 (95.5)	1467 (92.4)	0.12	1057 (94.6)	4132 (92.5)	0.08

History of acute MI	81 (20.4)	244 (15.4)	0.14	270 (24.2)	731 (16.4)	0.19

History of chronic alcohol use (3 yr)	7 (1.8)	19 (1.2)	0.05	29 (2.6)	69 (1.5)	0.08

History of chronic kidney disease (3 yr)	16 (4.0)	27 (1.7)	0.16	48 (4.3)	74 (1.7)	0.18

History of chronic liver disease (3 yr)	11 (2.8)	35 (2.2)	0.04	36 (3.2)	71 (1.6)	0.12

Residence in a long-term care facility	12 (3.0)	50 (3.1)	0.01	66 (5.9)	151 (3.4)	0.13

No. of prescription drugs in previous year, median (IQR)	15 (11–19)	12 (9–16)	0.41	15 (12–20)	12 (9–16)	0.55

Use of medications that increase risk of hemorrhage in preceding 120 days

ASA	≤ 5	6 (0.4)	0.02	14 (1.3)	24 (0.5)	0.09

ASA–dipyridamole	≤ 5	≤ 5	0.11	≤ 5	6 (0.1)	0.05

Clopidogrel	23 (5.8)	36 (2.3)	0.21	49 (4.4)	116 (2.6)	0.11

Warfarin	62 (15.6)	80 (5.0)	0.42	210 (18.8)	289 (6.5)	0.44

NSAIDs	32 (8.1)	122 (7.7)	0.01	98 (8.8)	368 (8.2)	0.02

Other medication use in preceding 120 days

β-adrenergic receptor antagonists	265 (66.8)	944 (59.4)	0.15	729 (65.3)	2667 (59.7)	0.11

ACE inhibitors	174 (43.8)	639 (40.2)	0.07	494 (44.2)	1862 (41.7)	0.05

Angiotensin-receptor blockers	100 (25.2)	356 (22.4)	0.07	251 (22.5)	995 (22.3)	0.00

Calcium-channel blockers	167 (42.1)	683 (43.0)	0.02	438 (39.2)	1811 (40.6)	0.03

Digoxin	71 (17.9)	275 (17.3)	0.01	209 (18.7)	717 (16.1)	0.07

Antiarrhythmics	22 (5.5)	143 (9.0)	0.13	101 (9.0)	357 (8.0)	0.04

Nitrates	53 (13.4)	153 (9.6)	0.12	184 (16.5)	428 (9.6)	0.22

P-glycoprotein inhibitors	38 (9.6)	146 (9.2)	0.01	114 (10.2)	360 (8.1)	0.08

P-glycoprotein inducers	14 (3.5)	21 (1.3)	0.17	27 (2.4)	65 (1.5)	0.08

Cytochrome P450 3A4 inhibitors	< 5	< 5	0.01	≤ 5	10 (0.2)	0.04

Formulation of dabigatran etexilate dispensed

110 mg only	288 (72.5)	1100 (69.3)	0.07	814 (72.9)	3098 (69.4)	0.08

150 mg only	93 (23.4)	407 (25.6)	0.05	252 (22.6)	1136 (25.4)	0.07

Both	16 (4.0)	81 (5.1)	0.05	51 (4.6)	231 (5.2)	0.03

Procedures in previous 5 years

Angiography	61 (15.4)	274 (17.3)	0.05	250 (22.4)	782 (17.5)	0.13

Carotid Doppler ultrasonography	209 (52.6)	469 (29.5)	0.50	386 (34.6)	1354 (30.3)	0.09

Carotid endarterectomy	6 (1.5)	10 (0.6)	0.10	7 (0.6)	18 (0.4)	0.03

Coronary artery bypass	8 (2.0)	40 (2.5)	0.03	44 (3.9)	137 (3.1)	0.05

Pacemaker insertion	25 (6.3)	178 (11.2)	0.16	120 (10.7)	434 (9.7)	0.03

Percutaneous coronary angioplasty	24 (6.0)	85 (5.4)	0.03	90 (8.1)	243 (5.4)	0.11

Valve surgery	12 (3.0)	54 (3.4)	0.02	56 (5.0)	174 (3.9)	0.06

Income quintile

1 (lowest)	87 (21.9)	290 (18.3)	0.09	225 (20.1)	777 (17.4)	0.07

2	74 (18.6)	328 (20.7)	0.05	253 (22.6)	868 (19.4)	0.08

3	80 (20.2)	330 (20.8)	0.02	200 (17.9)	874 (19.6)	0.04

4	82 (20.7)	311 (19.6)	0.03	212 (19.0)	904 (20.2)	0.03

5 (highest)	73 (18.4)	324 (20.4)	0.05	225 (20.1)	1030 (23.1)	0.074

Major hemorrhage in previous 5 years	–	–	–	222 (19.9)	441 (9.9)	0.31

Duration of statin use, median (IQR), yr	3.6 (1.1–7.0)	4.0 (1.9–7.9)	0.14	4.6 (1.8–8.5)	4.4 (1.9–7.9)	0.05

Note: ACE = angiotensin-converting-enzyme, ASA = acetylsalicylic acid, IQR = interquartile range, MI = myocardial infarction, NSAID = nonsteroidal anti-inflammatory drug, TIA = transient ischemic attack.
↵* Unless stated otherwise.
↵† Difference between cases and controls divided by standard deviation.

View popup

Table 3:

Association between statin use and stroke or transient ischemic attack

Statin	No. (%)		OR (95% CI)	Adjusted OR^* (95% CI)
Statin	Cases n = 397	Controls n = 1588	OR (95% CI)	Adjusted OR^* (95% CI)
Atorvastatin, pravastatin, fluvastatin or rosuvastatin (ref)	358 (90.2)	1438 (90.6)	1.00	1.00
Simvastatin or lovastatin	39 (9.8)	150 (9.4)	1.04 (0.72–1.51)	1.33 (0.88–2.01)

Note: CI = confidence interval, OR = odds ratio, ref = reference category.
↵* Adjusted for history of stroke or transient ischemic attack in preceding 5 years, Charlson Comorbidity Index score, diabetes, hypertension, myocardial infarction, congestive heart failure, chronic kidney disease, number of prescription drugs in previous year, medications (β-adrenergic receptor blockers, nitrates, acetylsalicylic acid and dipyridamole, antiarrhythmics, clopidogrel, warfarin, nonsteroidal anti-inflammatory drugs, P-glycoprotein inhibitors, cytochrome P450 3A4 inhibitors), carotid Doppler ultrasonography, pacemaker insertion and duration of statin use.

View popup

Table 4:

Association between statin use and major hemorrhage

Statin	No. (%)		OR (95% CI)	Adjusted OR^* (95% CI)
Statin	Cases n = 1117	Controls n = 4465	OR (95% CI)	Adjusted OR^* (95% CI)
Atorvastatin, pravastatin, fluvastatin or rosuvastatin (ref)	984 (88.1)	4069 (91.1)	1.00	1.00
Simvastatin or lovastatin	133 (11.9)	396 (8.9)	1.39 (1.13–1.71)	1.46 (1.17–1.82)

Note: CI = confidence interval, OR = odds ratio, ref = reference category.
↵* Adjusted for history of major hemorrhage, Charlson Comorbidity Index score, myocardial infarction, angina, congestive heart failure, chronic kidney disease, chronic liver disease, residence in long-term care facility, number of prescription drugs in previous year, medications (β-adrenergic receptor blockers, nitrates, clopidogrel, warfarin, nonsteroidal anti-inflammatory drugs, P-glycoprotein inhibitors, P-glycoprotein inducers, cytochrome P450 3A4 inhibitors), angiography and percutaneous coronary angioplasty.